How can I use this model with LLMPredictor() class from llama-index?

#15
by NastyaTima - opened

I want to train llama-2-7b-chat model on my pdf-documents using llama-index library. But after model initialization I always get "CUDA out of memory" error.
If I use it without LLMPredictor() everything is fine.
Please, tell me, how can I use it to ask questions based on my documents?

Huggingface Projects org

Hi @NastyaTima
I think it would be better to ask this question in the Forum or Discord.
I made this Space, but I'm not an expert on NLP or transformers, and don't know the answer.
I think you have a better chance of being seen by someone more knowledgeable if you ask your question there.

@hysts Hi
Okay, I got it, anyway thank you very much for answer :)

Sign up or log in to comment